Metal halide lamps began with the addition of the halides of various light-emitting metals to the high pressure memory lamp in order to modify its color and increase its operating efficacy as proposed by U.S. Pat. No. 3,234,421--Reiling, issued in 1966. Since then metal halide lamps have become commercially useful for general illumination; their construction and mode of operation are described in IES Lighting Handbook, 5th Edition, 1972, published by the Illuminating Engineering Society, pages 8-34.
The metal halide lamp generally operates with a substantially fully vaporized charge of mercury and an unvaporized excess consisting mostly of metal iodides in liquid form. The favored filling comprises the iodides of sodium, scandium and thorium. The operating conditions together with the geometrical design of the lamp envelope must provide sufficiently high temperatures, particularly in the ends, to vaporize a substantial quantity of the iodides, especially of the NaI. In general, this requires minimum temperatures under operating conditions of the order of 700.degree. C.
The quantity of a metal salt which may be accommodated in the vapor state within a given volume at a given temperature, for instance NaI at 750.degree. C., can be readily calculated. However, the metal salt charge, and in particular the charge of NaI, that is put into metal halide lamps of commercial manufacture is usually many times greater than such calculated quantity. While most of the added NaI remains as condensate within the arc tube, the quantity participating in the arc discharge does increase with the total quantity put into the tube, even though at a diminishing rate, and serves to improve the efficacy and lower the color temperature of the lamp.
The desirability of having the excess metal halide widely distributed rather than condensed in the ends of the lamp is known. To achieve this result, it has been proposed to design the arc tube in such a way that the end temperatures are higher than that in the middle, so that excess metal iodide will tend to condense about the middle of the arc tube. However in the usual quartz or fused silica arc tube, the condensate does not form a true film in the sense of a continuous layer on the inside surface of the envelope, but tends to remain as discrete droplets. In the aforementioned Bateman application, one means disclosed for promoting the formation and spreading of a liquid condensate film is a coating of fine particles of refractory material on the interior surface of the envelope. Such a coating may be formed by contacting the inside of the envelope by a silica smoke which is then partially sintered to compact it into a more rugged structure and improve its adherence. The coating provides a surface which causes the condensate to spread out into a film.